1 2 3 4 5 6 7 



YEAR 



Figure 2.— Predicted versus observed tree ioss by 

 year based on q^, by grouped tree diameter 

 classes and total stand (Situation A. Klein and 

 others 1978). 



1 2 3 4 5 6 7 



YEAR 



Figure 3.— Predicted versus ob- 

 served tree loss by year for total 

 stand based on (Situation A, 

 Klein and others 1978). 



The second data set came from a mountain pine beetle 

 infestation in the Gallatin River Drainage, and was used 

 only for total tree loss, because the data were not 

 originally taken by diameter classes (situation B). Tree 

 loss over time did not fall into the usual bell-shaped pat- 

 tern, yet the predicted tree loss approximated the actual 

 double-peaked curve (table 4; fig. 4) (Burnell 1977). 



Answers to the questions, "Which of the lodgepole 

 pine stands are the most susceptible to mountain pine 

 beetle outbreak development?" and "How many trees 

 will the manager lose if the stand becomes infested?" de- 

 pend upon risk. A definition of risk has two parts: (1) 

 probability of an outbreak within a set time period, and 

 (2) expected loss in the advent of an outbreak (Safranyik 

 1982). Rehable methods are not available to predict 

 when an outbreak wiU develop, but we can predict the 

 most susceptible stands and also forecast stand deple- 

 tion in terms of stand structure should an epidemic oc- 

 cur. To date, six risk-rating systems have been 

 developed that are based on climatic and tree/stand 

 variables having a major effect on beetle survival and 

 distribution. Rate of spread could be considered using 

 historical maps (fig. 5) or mathematical models based on 

 habitat type. 



8 



